CN109284363A - A kind of answering method, device, electronic equipment and storage medium - Google Patents
A kind of answering method, device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN109284363A CN109284363A CN201811465133.0A CN201811465133A CN109284363A CN 109284363 A CN109284363 A CN 109284363A CN 201811465133 A CN201811465133 A CN 201811465133A CN 109284363 A CN109284363 A CN 109284363A
- Authority
- CN
- China
- Prior art keywords
- question
- retrieval
- target
- processed
- matched
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 93
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 65
- 238000013507 mapping Methods 0.000 claims abstract description 19
- 238000012545 processing Methods 0.000 claims description 18
- 238000012216 screening Methods 0.000 claims description 13
- 230000001502 supplementing effect Effects 0.000 claims description 11
- 238000004891 communication Methods 0.000 claims description 4
- 238000000605 extraction Methods 0.000 abstract description 7
- 239000013589 supplement Substances 0.000 abstract description 7
- 238000010586 diagram Methods 0.000 description 17
- 230000008569 process Effects 0.000 description 15
- 238000004590 computer program Methods 0.000 description 9
- 230000008901 benefit Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013145 classification model Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 230000008909 emotion recognition Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000004807 localization Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000000844 transformation Methods 0.000 description 1
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the invention discloses a kind of answering method, device, electronic equipment and storage mediums, which comprises the problem of being inputted using natural language understanding algorithm to user carries out information extraction and supplement, obtains problem to be processed；The problem of is carried out by Question Classification, obtains the problem to be processed for the problem to be processed type；Reasoning algorithm mapping relations are retrieved according to preset problem types and map, are obtained and the matched search result of problem to be processed for described problem type using matched map retrieval reasoning algorithm；The search result is handled by default rule template and/or spatial term method, generate the matched answer of described problem and returns to the user.The technical solution of the embodiment of the present invention can be improved the answer accuracy rate and efficiency of question answering system.
Description
Technical Field
The embodiment of the invention relates to the technical field of natural language processing, in particular to a question answering method, a question answering device, electronic equipment and a storage medium.
Background
With the popularization of natural language understanding technology, many intelligent question-answering systems have appeared. Through establishing a large amount of knowledge bases, the sentences in the natural language are segmented and labeled, and then the segmented sentences are compared with the knowledge bases which are trained through a large amount of corpora, so that the machine can understand the natural language. The question-answering system can collect huge internet information, gives answers to problems of life or work and the like of users, and provides convenience for life of the information era of people. The intelligent question-answering system can directly talk with the user, dig out information in user dialogue, and accurately judge the intention of the user in question, so that answers meeting the intention of the user can be intelligently given.
For question-answering systems, users often present some particularly complex queries. Conventional question-answering systems typically only perform simple processing transformations on questions entered by a user and retrieve corresponding data from a predefined knowledge base based on the resulting keywords, in a manner very similar to existing search engines.
In the specific implementation process, the inventor finds that the following problems exist in the prior art: the traditional question-answering system usually does not do semantic understanding to obtain the intention of the user, so that the relevance of the searched answer and the question is poor, and the returned result is probably not required by the user at all. Meanwhile, the traditional question-answer interaction system is based on knowledge positioning of a plain text, often comprises data structures such as an inverted list and the like, and needs comprehensive ranking of the inverted list of a plurality of keywords, so that the answer output efficiency is low.
Disclosure of Invention
In view of this, embodiments of the present invention provide a question answering method, device, electronic device and storage medium, and mainly aim to solve the problem of low answer accuracy and efficiency of a question answering system.
In order to solve the above problems, embodiments of the present invention mainly provide the following technical solutions:
in a first aspect, an embodiment of the present invention provides a question answering method, including:
extracting and supplementing information of the problems input by the user by adopting a natural language understanding algorithm to obtain the problems to be processed;
performing problem classification on the problems to be processed to obtain the problem types of the problems to be processed;
according to a preset mapping relation between the problem type and a map retrieval inference algorithm, aiming at the problem type, adopting a matched map retrieval inference algorithm to obtain a retrieval result matched with the problem to be processed;
and processing the retrieval result through a preset rule template and/or a natural language generation method, generating an answer matched with the question and returning the answer to the user.
In a second aspect, an embodiment of the present invention further provides a question answering device, where the question answering device includes:
the problem acquisition module is used for extracting and supplementing information of the problems input by the user by adopting a natural language understanding algorithm to obtain the problems to be processed;
the problem classification module is used for performing problem classification on the problems to be processed to obtain the problem types of the problems to be processed;
the retrieval result acquisition module is used for acquiring a retrieval result matched with the problem to be processed by adopting a matched graph retrieval inference algorithm aiming at the problem type according to a preset mapping relation between the problem type and the graph retrieval inference algorithm;
and the answer generation module is used for processing the retrieval result through a preset rule template and/or a natural language generation method, generating an answer matched with the question and returning the answer to the user.
In a third aspect, an embodiment of the present invention further provides an electronic device, including:
at least one processor;
and at least one memory, bus connected with the processor; wherein,
the processor and the memory complete mutual communication through the bus;
the processor is used for calling the program instructions in the memory so as to execute the question answering method provided by any embodiment of the invention.
In a fourth aspect, an embodiment of the present invention further provides a non-transitory computer-readable storage medium, which stores computer instructions, where the computer instructions cause the computer to execute the question answering method provided in any embodiment of the present invention.
By the technical scheme, the technical scheme provided by the embodiment of the invention at least has the following advantages:
according to the question answering method provided by the embodiment of the invention, the natural language understanding algorithm is adopted to extract and supplement information of the questions input by the user to obtain the questions to be processed, the questions are classified to obtain the corresponding question types, then the matched graph retrieval reasoning algorithm is adopted to obtain the retrieval result matched with the questions to be processed according to the mapping relation between the preset question types and the graph retrieval reasoning algorithm, and finally the retrieval result is processed through the preset rule template and/or the natural language generating method, so that the final answer is generated and returned to the user, the problem that the existing question answering system is low in answer accuracy and efficiency is solved, and the answer accuracy and efficiency of the question answering system are improved.
The foregoing description is only an overview of the technical solutions of the embodiments of the present invention, and the embodiments of the present invention can be implemented according to the content of the description in order to make the technical means of the embodiments of the present invention more clearly understood, and the detailed description of the embodiments of the present invention is provided below in order to make the foregoing and other objects, features, and advantages of the embodiments of the present invention more clearly understandable.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the embodiments of the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:
fig. 1 is a flowchart of a question answering method according to an embodiment of the present invention;
fig. 2a is a flowchart of a question answering method according to a second embodiment of the present invention;
fig. 2b is a flowchart of a question answering method according to a second embodiment of the present invention;
FIG. 2c is a diagram illustrating a process of retrieving a map according to a second embodiment of the present invention;
FIG. 2d is a schematic diagram of a portion of financial intellectual entities and relationships in a financial intellectual map stored in a map database according to a second embodiment of the present invention;
fig. 3 is a schematic diagram of a question answering device according to a third embodiment of the present invention;
fig. 4 is a schematic structural diagram of an electronic device according to a fourth embodiment of the present invention.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
Example one
Fig. 1 is a flowchart of a question answering method according to an embodiment of the present invention, which is applicable to a case where answers to questions are output quickly and accurately. Accordingly, as shown in fig. 1, the method comprises the following operations:
and S110, extracting and supplementing information of the problem input by the user by adopting a natural language understanding algorithm to obtain the problem to be processed.
The natural language understanding algorithm may be any method for realizing effective communication between a person and a computer by using natural language, such as a series of methods of keyword matching, question keyword extraction, dictionaries, synonyms, named entity recognition, rule-based sentence matching or syntactic dependency analysis. The problem to be processed may be a problem formed after data processing. The data processing includes, but is not limited to, processes of traditional and simple conversion, Chinese word segmentation, part of speech tagging, data cleansing, syntax parsing, entity recognition and/or speech-to-text conversion.
In the embodiment of the invention, the user can input questions to the question answering system by adopting modes of manual input, voice input and the like. After the question-answering system obtains the questions input by the user, the question-answering system can extract and supplement information for the questions by adopting a natural language understanding algorithm. The information extraction method may include, but is not limited to, extraction based on part-of-speech tagging, extraction based on semantic analysis, extraction based on piece-of-speech analysis, and the like. The problem is supplemented mainly by detecting and completing the missing part of the sentence structure in the problem proposed by the user. After information extraction and supplementary processing, the question answering system can obtain the questions to be processed. The process of extracting and supplementing the information of the question is also the process of understanding the question. In the process of understanding the question, the question-answering system can judge the intention of the user.
S120, performing problem classification on the problems to be processed to obtain the problem types of the problems to be processed.
Correspondingly, after the to-be-processed problem is obtained, the question answering system can classify the to-be-processed problem to obtain the problem type corresponding to the to-be-processed problem. Optionally, the classification model may be used to classify the problem to be processed from multiple aspects such as problem type, user behavior, emotion recognition, and the like. Accordingly, the question types may include conceptual class questions, non-conceptual class questions, and the like.
S130, according to a preset mapping relation between the problem types and the map retrieval inference algorithm, aiming at the problem types, adopting a matched map retrieval inference algorithm to obtain retrieval results matched with the problems to be processed.
The graph retrieval reasoning algorithm can be a method applied to the knowledge graph and used for solving question answering and information retrieval.
Knowledge-graphs are collections of large amounts of knowledge. At present, the expression form of knowledge is generally a large amount of unstructured text, and a knowledge graph is a form that a large amount of text containing knowledge is changed into a triple, namely an entity-relationship-entity (entity), and is stored and presented in a certain way, so that some retrieval methods are provided for people to acquire knowledge. Meanwhile, the knowledge map can be effectively subjected to knowledge expansion by combining an inference technology, and the coverage of knowledge is increased. The knowledge graph constructs a knowledge base-based method to provide a certain degree of knowledge and knowledge association for the computer, and in this form, the computer performs intention judgment on the problems of the user according to knowledge base information. The knowledge base is established by means of network resources and comprises related field entity words, professional entity words and corresponding entity word attributes. Further, associations (i.e., relationships) between these entity words may be established from the question-answer dataset, thereby revealing the associations between the entity words using a knowledge-relationship graph.
Applying knowledge maps to a question-answering system has the following advantages: 1) semantic understanding is more intelligent. The semantic understanding degree is a core index of the question-answering system. For plain text data, semantic understanding is often built on similarity calculation of question sentences and text sentences. The essence of semantic understanding and knowledge, however, is that the association, such a one-to-one similarity calculation, ignores the data association. In a knowledge graph, all knowledge points are associated by edges with semantic information. In the matching and associating process from the question to the knowledge point of the knowledge graph, a large amount of associated information of associated nodes can be used, and the associated information undoubtedly provides conditions for intelligent semantic understanding. 2) The answer accuracy is higher. The knowledge of the knowledge map comes from professional annotation or formatting capture of a professional database, so that the high accuracy of the data is fundamentally ensured. In the plain text, since the same kind of knowledge is easy to be mentioned in the text for many times, the phenomenon of inconsistent data is caused, and the accuracy rate of the knowledge is reduced. 3) The retrieval efficiency is higher. The structured organization form of the knowledge graph provides format support for the quick knowledge retrieval of the computer. The computer can use a Structured Language such as SQL (Structured Query Language), sparql (simple Protocol and rdf Query Language), etc. to perform accurate knowledge localization. For knowledge positioning of plain text, the knowledge positioning often includes data structures such as inverted lists, comprehensive ranking of the inverted lists of a plurality of keywords is needed, and efficiency is low.
In the embodiment of the invention, aiming at different problem types of the problems to be processed, different map retrieval inference algorithms can be adopted to retrieve in the knowledge map so as to obtain the retrieval result matched with the problems to be processed. Therefore, after the problem type corresponding to the problem to be processed is obtained, the matched graph retrieval inference algorithm can be determined for the problem type according to the mapping relation of the problem type and the graph retrieval inference algorithm.
And S140, processing the retrieval result through a preset rule template and/or a natural language generation method, generating an answer matched with the question and returning the answer to the user.
The rule template and the natural language generating method are both methods for generating natural texts by processing retrieval results obtained by using the knowledge graph. For example, the rule template may be a heuristic rule, the natural language generation method may be a Beam Search (Beam Search), a Random Search (Random Search), and the like, and the specific content of the rule template and the natural language generation method is not limited in the embodiments of the present invention.
It can be understood that the retrieval result obtained in the knowledge graph by using the graph retrieval inference algorithm only comprises entity and relationship information. Therefore, after the retrieval result is obtained in the knowledge graph by using the graph retrieval inference algorithm, the retrieval result can be organized and processed by a preset rule template and/or a natural language generation method to obtain a corresponding natural text as an answer for question matching, and the finally generated answer is returned to the user.
According to the question answering method provided by the embodiment of the invention, the natural language understanding algorithm is adopted to extract and supplement information of the questions input by the user to obtain the questions to be processed, the questions are classified to obtain the corresponding question types, then the matched graph retrieval reasoning algorithm is adopted to obtain the retrieval result matched with the questions to be processed according to the mapping relation between the preset question types and the graph retrieval reasoning algorithm, and finally the retrieval result is processed through the preset rule template and/or the natural language generating method, so that the final answer is generated and returned to the user, the problem that the existing question answering system is low in answer accuracy and efficiency is solved, and the answer accuracy and efficiency of the question answering system are improved.
Example two
Fig. 2a is a flowchart of a question-answering method according to a second embodiment of the present invention, and fig. 2b is a flowchart of a question-answering method according to a second embodiment of the present invention, which is embodied based on the above embodiments, in this embodiment, a specific type of a question type is given, and a specific implementation manner of obtaining a search result matched with the question to be processed by using a matching atlas retrieval inference algorithm for the question type according to a preset mapping relationship between the question type and the atlas retrieval inference algorithm is given, and a construction manner of a preset knowledge atlas is given at the same time. Accordingly, as shown in fig. 2a and 2b, the method comprises the following operations:
s210, extracting and supplementing information of the problem input by the user by adopting a natural language understanding algorithm to obtain the problem to be processed.
S220, performing problem classification on the problems to be processed to obtain the problem types of the problems to be processed.
And S230, according to a preset mapping relation between the problem types and the map retrieval inference algorithm, aiming at the problem types, adopting a matched map retrieval inference algorithm to obtain retrieval results matched with the problems to be processed.
Wherein the problem types include concept class problems and non-concept class problems. Accordingly, S230 may specifically include the following two sets of operations S231a-S232a and S231b-S233b in parallel.
Accordingly, when the question type is a concept-class question, S230 may specifically include operations S231a-S232a, and the like.
S231a, extracting concept subject information of the problem to be processed as a target entity, and supplementing a target relation corresponding to the target entity.
Correspondingly, if the question answering system determines that the question type of the question to be processed is a concept type question, concept subject information in the question can be directly provided as a target entity, and a target relation corresponding to the target entity is supplemented.
Illustratively, suppose the question to be processed is "what is RMB? "then its corresponding target entity may be" RMB "and the corresponding target relationship may be" Definitions ".
S232a, retrieving in a preset knowledge graph according to the target entity and the target relation to obtain the retrieval result.
The preset knowledge graph can be a knowledge graph constructed in any professional field. Such as the financial field or the entertainment news field, the embodiments of the present invention do not limit the fields related to the preset knowledge graph and the specific content included in the preset knowledge graph.
In the embodiment of the invention, after the target entity and the target relationship corresponding to the concept problem are determined, the target entity and the target relationship can be directly mapped to the preset knowledge graph for retrieval.
Accordingly, when the question type is a non-conceptual type question, S230 may specifically include operations of S231b-S233b, and the like.
S231b, filling entity and relation information of the problem to be processed to obtain a list comprising at least one group of intermediate entities and intermediate relations.
In the embodiment of the invention, for non-conceptual problems, accurate entities and relations may not be obtained through problem understanding and classification. At this point, the question-answering system may perform entity and relationship information population on the question to be processed, thereby obtaining a list including at least one set of intermediate entities and intermediate relationships. A list including sets of intermediate entities and intermediate relationships may then be retrieved and answer output.
S232b, screening the intermediate entities and the intermediate relations in the list to obtain target entities and target relations.
Correspondingly, after the list is obtained, the intermediate entities and the intermediate relationships in the list can be screened to obtain the target entities and the target relationships. It should be noted that the finally obtained target entities and target relationships may be one group or multiple groups, and the specific needs are determined according to the contents of the problems to be processed.
For example, assuming that the pending question is "what types the role of the clique loan service specifically includes," its corresponding target entities and target relationships may be a set. Assuming that the problem to be processed is "what types the role of the banking loan service specifically includes and what content the specific responsibilities of the banking member include", there may be at least two groups of target entities and target relationships that are finally obtained, corresponding to the sub-problem "what types the role of the banking loan service specifically includes" and the sub-problem "what content the specific responsibilities of the banking member includes" respectively.
S233b, retrieving in a preset knowledge graph according to the target entity and the target relation, and obtaining the retrieval result.
Similarly, in the embodiment of the present invention, after the target entity and the target relationship corresponding to the non-concept problem are determined, the target entity and the target relationship may be mapped to the preset knowledge graph for retrieval.
In an optional embodiment of the present invention, the preset knowledge map is stored in a map database;
retrieving in a preset knowledge graph according to the target entity and the target relationship to obtain the retrieval result, which may include:
performing accurate retrieval in a preset knowledge graph according to the complete name of the target entity and the target relation to obtain a target subgraph as the retrieval result; or
And carrying out fuzzy retrieval in a preset knowledge graph according to the target entity and the partial name of the target relation to obtain at least two groups of matched associated subgraphs as the retrieval result.
In the embodiment of the invention, the preset knowledge graph can be stored in a graph database mode. Meanwhile, the preset knowledge graph can provide various retrieval modes. The first is an accurate retrieval mode, that is, an accurate retrieval based on entity and relationship names. Specifically, the complete name of the target entity and the target relationship to be queried can be input in the preset knowledge graph, so that accurate query is performed in the preset knowledge graph according to the target entity and the target relationship, and a retrieval result can be returned in a sub-graph mode. Meanwhile, the accurate retrieval can also support the specified retrieval depth so as to conveniently inquire more and deeper knowledge. The second is a fuzzy retrieval mode, namely fuzzy retrieval based on entity and relation name. Specifically, partial names of target entities to be queried and target relationships can be input in a preset knowledge graph, a retrieval engine matched with the preset knowledge graph can score multiple acquired retrieval results by adopting a similarity matching algorithm, and the associated subgraphs of the first matched entity-relationships are returned as the retrieval results according to the scoring results. It should be noted that the preset knowledge graph can also provide an inquiry type retrieval mode based on a natural language mode. The query type retrieval mode does not need a question-answering system to acquire a target entity and a target relation in advance according to question classification, a retrieval engine matched with a preset knowledge graph can realize basic natural language understanding algorithms such as question understanding, question classification and intention identification, further maps knowledge contained in the question to the entity and the relation of the preset knowledge graph, performs graph retrieval based on the accurate retrieval or fuzzy retrieval mode, and returns a retrieval result.
Fig. 2c is a diagram of a map retrieval process according to a second embodiment of the present invention, and as shown in fig. 2c, the three retrieval methods may be summarized as follows: after the retrieval content is input in the preset knowledge graph, fuzzy matching query can be firstly carried out. If the matching threshold is determined to be reached, the processes of entity mapping, problem classification, rule matching, map retrieval and the like can be carried out, otherwise, the operations of word segmentation, entity identification, keyword identification and the like are carried out on the input retrieval content, and finally, the retrieval result is output in a sub-graph mode after the subsequent links of entity mapping and the like are entered. It should be noted that the knowledge is not invariable, some knowledge may have timeliness, and the preset knowledge map is difficult to achieve 100% accuracy in the automatic construction process. Therefore, the management system for presetting the knowledge graph can support editing operations such as increasing, deleting, modifying, checking and the like on the entity and the relation of the graph in a visual mode, and meanwhile, the retrieval modes can also support interaction in a visual mode.
S240, processing the retrieval result through a preset rule template and/or a natural language generating method to generate an answer matched with the question.
S250, judging whether the association degree between the target entity and the answer matched with the target relation and the question meets the requirement of a preset association degree threshold value through a preset reasoning algorithm, and if so, executing S270; otherwise, return to execution S232b, or execution S260.
The preset inference algorithm may be an inference algorithm adopted by the preset knowledge graph, such as a knowledge graph inference algorithm of path tensor decomposition, and the embodiment of the present invention does not limit the specific type of the preset inference algorithm. The preset relevance threshold may be a threshold set according to actual requirements, and the embodiment of the present invention does not limit a specific numerical value of the preset relevance threshold.
It should be noted that, if it is determined that the question type is a non-conceptual question, after the search result is processed by a preset rule template and/or a natural language generation method to generate an answer for question matching, the association degree between the screened target entity and the target relationship and the answer for question matching needs to be judged by a preset inference algorithm. When the relevance does not meet the requirement of the preset relevance threshold, for example, the relevance is lower than the preset relevance threshold, the relevance between the retrieval result and the question is low, and the preset knowledge graph can be reused for retrieval or answers matched with the question can be input in other modes.
When the preset knowledge graph is reused for retrieval, the operation of screening the intermediate entities and the intermediate relations in the list to obtain the target entities and the target relations can be returned to be executed. Optionally, the screening method at this time may include two types: the first is to further screen the target entity and the target relationship obtained in the last screening to obtain a new target entity and a new target relationship. For example, if 3 sets of target entities and target relationships are obtained from the previous screening, new target entities and target relationships may be obtained from the 3 sets of target entities and target relationships by further screening. And the second method still utilizes the intermediate entities and the intermediate relations in the initially formed list to carry out screening to obtain new target entities and target relations. For example, the initially obtained list includes 5 sets of intermediate entities and intermediate relationships, and the first filtering obtains 2 sets of target entities and target relationships. And when the re-screening is returned, the screening is still carried out according to the list which is initially obtained and comprises 5 groups of intermediate entities and intermediate relations, and 2 groups of new target entities and target relations are obtained.
And S260, generating answers matched with the questions directly through a preset rule template and/or a natural language generation method.
Correspondingly, in the embodiment of the invention, when the relevance does not meet the requirement of the preset relevance threshold, the answer matched with the question can be generated directly through a preset rule template and/or a natural language generation method.
And S270, returning the answer matched with the answer question to the user.
In an optional embodiment of the invention, the preset knowledge-graph is pre-constructed based on financial management regulations.
In the embodiment of the present invention, optionally, the preset knowledge graph may be pre-constructed based on financial management regulations. Specifically, the relevant learning data of the financial management regulation can be prepared in advance, the relevant learning data of the financial management regulation can be learned by means of a machine learning method, and the learned financial field knowledge entities and the relations between the financial field knowledge entities are sorted and stored in a database to obtain a matched database. FIG. 2d is a schematic diagram of a portion of financial intellectual entities and relationships in a financial intellectual map stored in a map database according to a second embodiment of the present invention. As shown in fig. 2d, the financial knowledge graph may visually display the related information in the graph according to the query condition, and by clicking the node corresponding to the entity or the contact corresponding to the relationship, an operable prompt may be provided to complete expansion and modification of the knowledge graph information.
It should be noted that fig. 2a and fig. 2b are only schematic diagrams of an implementation manner, and the execution sequence between S231a-S232a and S231b-S233b is not sequential. That is, S231a-S232a may be executed first, and then S231b-S233b may be executed first, or S231b-S233b may be executed first, and then S231a-S232a may be executed second, or both may be executed alternatively or in parallel.
By adopting the technical scheme, the matched retrieval result is obtained by adopting the matched map retrieval reasoning algorithm according to the specific question type determined by the user question, the answer matched with the question is generated according to the retrieval result and returned to the user, the problem of low answer accuracy and efficiency of the existing question-answering system is solved, and the answer accuracy and efficiency of the question-answering system are improved.
It should be noted that any permutation and combination between the technical features in the above embodiments also belong to the scope of the present invention.
EXAMPLE III
Fig. 3 is a schematic diagram of a question answering device according to a third embodiment of the present invention, and as shown in fig. 3, the question answering device includes: a question obtaining module 310, a question classifying module 320, a retrieval result obtaining module 330, and an answer generating module 340, wherein:
the problem acquisition module 310 is configured to extract and supplement information for a problem input by a user by using a natural language understanding algorithm to obtain a problem to be processed;
the problem classification module 320 is configured to perform problem classification on the problem to be processed to obtain a problem type of the problem to be processed;
the retrieval result obtaining module 330 is configured to obtain a retrieval result matched with the problem to be processed by using a matched graph retrieval inference algorithm for the problem type according to a preset mapping relationship between the problem type and the graph retrieval inference algorithm;
and the answer generating module 340 is configured to process the search result through a preset rule template and/or a natural language generating method, generate an answer matched with the question, and return the answer to the user.
According to the embodiment of the invention, the natural language understanding algorithm is adopted to extract and supplement information of the questions input by the user to obtain the questions to be processed, the questions are classified to obtain corresponding question types, then the matched graph retrieval reasoning algorithm is adopted to obtain the retrieval result matched with the questions to be processed according to the mapping relation between the preset question types and the graph retrieval reasoning algorithm, and finally the retrieval result is processed through the preset rule template and/or the natural language generating method, so that the final answer is generated and returned to the user, the problems of low answer accuracy and efficiency of the existing question-answering system are solved, and the answer accuracy and efficiency of the question-answering system are improved.
Optionally, the question type includes a concept class question; a retrieval result obtaining module 330, configured to specifically extract concept subject information of the problem to be processed as a target entity, and supplement a target relationship corresponding to the target entity; and retrieving in a preset knowledge graph according to the target entity and the target relation to obtain the retrieval result.
Optionally, the question type includes a non-conceptual class question; a retrieval result obtaining module 330, configured to perform entity and relationship information filling on the to-be-processed problem to obtain a list including at least one group of intermediate entities and intermediate relationships; screening the intermediate entities and the intermediate relations in the list to obtain target entities and target relations; and retrieving in a preset knowledge graph according to the target entity and the target relation to obtain the retrieval result.
Optionally, the apparatus further comprises: the association degree determining module is used for judging the association degree between the target entity and the answer matched with the target relation and the question through a preset reasoning algorithm; the return execution module is used for returning and executing the operation of screening the intermediate entities and the intermediate relations in the list to obtain the target entities and the target relations if the relevance does not meet the requirement of a preset relevance threshold; and the first answer returning module is used for returning the answer matched with the answer question to the user.
Optionally, the apparatus further comprises: and the second answer returning module is used for directly generating an answer matched with the question through a preset rule template and/or a natural language generating method and returning the answer to the user if the relevance does not meet the requirement of a preset relevance threshold.
Optionally, the preset knowledge graph is stored in a graph database mode; a retrieval result obtaining module 330, configured to perform accurate retrieval in a preset knowledge graph according to the complete name of the target entity and the target relationship, and obtain a target sub-graph as the retrieval result; or
And carrying out fuzzy retrieval in a preset knowledge graph according to the target entity and the partial name of the target relation to obtain at least two groups of matched associated subgraphs as the retrieval result.
Optionally, the preset knowledge graph is pre-constructed based on financial management regulations.
The question answering device can execute the question answering method provided by any embodiment of the invention, and has corresponding functional modules and beneficial effects of the execution method. For technical details that are not described in detail in this embodiment, reference may be made to the question answering method provided in any embodiment of the present invention.
Since the question answering device described in this embodiment is a device capable of executing the question answering method in the embodiment of the present invention, based on the question answering method described in the embodiment of the present invention, those skilled in the art can understand the specific implementation manner of the question answering device of this embodiment and various variations thereof, and therefore, how to implement the question answering method in the embodiment of the present invention by the question answering device is not described in detail herein. The device used by those skilled in the art to implement the question answering method in the embodiments of the present invention is within the scope of the present application.
Example four
Fig. 4 is a schematic structural diagram of an electronic device according to a fourth embodiment of the present invention. As shown in fig. 4, the electronic apparatus includes: at least one processor (processor) 41; and at least one memory (memory)42, a bus 43 connected to the processor 41; wherein,
the processor 41 and the memory 42 complete mutual communication through the bus 43;
the processor 41 is configured to call program instructions in the memory 42 to perform the steps in the above-described embodiment of the question-answering method. For example, the processor 41 performs: extracting and supplementing information of the problems input by the user by adopting a natural language understanding algorithm to obtain the problems to be processed; performing problem classification on the problems to be processed to obtain the problem types of the problems to be processed; according to a preset mapping relation between the problem type and a map retrieval inference algorithm, aiming at the problem type, adopting a matched map retrieval inference algorithm to obtain a retrieval result matched with the problem to be processed; and processing the retrieval result through a preset rule template and/or a natural language generation method, generating an answer matched with the question and returning the answer to the user.
EXAMPLE five
An embodiment five of the present invention provides a non-transitory computer-readable storage medium, where the non-transitory computer-readable storage medium stores a computer instruction, and the computer instruction causes the computer to execute the question answering method provided in each of the above method embodiments: extracting and supplementing information of the problems input by the user by adopting a natural language understanding algorithm to obtain the problems to be processed; performing problem classification on the problems to be processed to obtain the problem types of the problems to be processed; according to a preset mapping relation between the problem type and a map retrieval inference algorithm, aiming at the problem type, adopting a matched map retrieval inference algorithm to obtain a retrieval result matched with the problem to be processed; and processing the retrieval result through a preset rule template and/or a natural language generation method, generating an answer matched with the question and returning the answer to the user.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, Compact-Read Only Memory (CD-ROM), optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The Memory may include volatile Memory in a computer readable medium, Random Access Memory (RAM), and/or nonvolatile Memory such as Read Only Memory (ROM) or flash Memory (flash RAM). The memory is an example of a computer-readable medium.
Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change Memory (PRAM), Static Random-Access Memory (SRAM), Dynamic Random-Access Memory (DRAM), other types of Random-Access Memory (RAM), Read-only Memory (ROM), Electrically Erasable Programmable Read-only Memory (EEPROM), flash Memory or other Memory technology, compact Disc Read-only Memory (CD-ROM), Digital Versatile Disks (DVD) or other optical storage, magnetic cassettes, magnetic tape storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in the process, method, article, or apparatus that comprises the element.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The above are merely examples of the present application and are not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
Claims (10)
1. A question-answering method, comprising:
extracting and supplementing information of the problems input by the user by adopting a natural language understanding algorithm to obtain the problems to be processed;
performing problem classification on the problems to be processed to obtain the problem types of the problems to be processed;
according to a preset mapping relation between the problem type and a map retrieval inference algorithm, aiming at the problem type, adopting a matched map retrieval inference algorithm to obtain a retrieval result matched with the problem to be processed;
and processing the retrieval result through a preset rule template and/or a natural language generation method, generating an answer matched with the question and returning the answer to the user.
2. The method of claim 1, wherein the question types include concept class questions;
according to a preset mapping relation between the problem type and a map retrieval inference algorithm, aiming at the problem type, adopting a matched map retrieval inference algorithm to obtain a retrieval result matched with the problem to be processed, wherein the method comprises the following steps:
extracting concept main body information of the problem to be processed as a target entity, and supplementing a target relation corresponding to the target entity;
and retrieving in a preset knowledge graph according to the target entity and the target relation to obtain the retrieval result.
3. The method of claim 1, wherein the question types include non-conceptual class questions;
according to a preset mapping relation between the problem type and a map retrieval inference algorithm, aiming at the problem type, adopting a matched map retrieval inference algorithm to obtain a retrieval result matched with the problem to be processed, wherein the method comprises the following steps:
filling entity and relation information of the problem to be processed to obtain a list comprising at least one group of intermediate entities and intermediate relations;
screening the intermediate entities and the intermediate relations in the list to obtain target entities and target relations;
and retrieving in a preset knowledge graph according to the target entity and the target relation to obtain the retrieval result.
4. The method according to claim 3, further comprising, after processing the search result by a preset rule template and/or a natural language generation method to generate an answer for the question matching:
judging the association degree between the target entity and the answer matched with the target relation and the question through a preset reasoning algorithm;
if the association degree is determined not to meet the requirement of a preset association degree threshold, returning to execute the operation of screening the intermediate entities and the intermediate relations in the list to obtain the operation of the target entities and the target relations until the association degree meets the requirement of the preset association degree threshold;
and returning the answer matched with the answer question to the user.
5. The method according to claim 4, after determining the relevance between the target entity and the answer matched with the target relation and the question through a preset inference algorithm, further comprising:
and if the relevance does not meet the requirement of a preset relevance threshold, generating an answer matched with the question directly through a preset rule template and/or a natural language generation method and returning the answer to the user.
6. The method according to any one of claims 2 or 3, wherein the predetermined knowledge-graph is stored using a graph database;
retrieving in a preset knowledge graph according to the target entity and the target relation to obtain the retrieval result, wherein the retrieving comprises:
performing accurate retrieval in the preset knowledge graph according to the complete name of the target entity and the target relation to obtain a target subgraph as the retrieval result; or
And carrying out fuzzy retrieval in the preset knowledge graph according to the target entity and the partial name of the target relation to obtain at least two groups of matched associated subgraphs as the retrieval result.
7. The method of any of claims 2-5, wherein the predetermined knowledge-graph is pre-constructed based on financial management regulations.
8. A question answering device, comprising:
the problem acquisition module is used for extracting and supplementing information of the problems input by the user by adopting a natural language understanding algorithm to obtain the problems to be processed;
the problem classification module is used for performing problem classification on the problems to be processed to obtain the problem types of the problems to be processed;
the retrieval result acquisition module is used for acquiring a retrieval result matched with the problem to be processed by adopting a matched graph retrieval inference algorithm aiming at the problem type according to a preset mapping relation between the problem type and the graph retrieval inference algorithm;
and the answer generation module is used for processing the retrieval result through a preset rule template and/or a natural language generation method, generating an answer matched with the question and returning the answer to the user.
9. An electronic device, comprising:
at least one processor;
and at least one memory, bus connected with the processor; wherein,
the processor and the memory complete mutual communication through the bus;
the processor is configured to call program instructions in the memory to perform the question answering method of any one of claims 1 to 7.
10. A non-transitory computer-readable storage medium storing computer instructions for causing a computer to perform the question-answering method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811465133.0A CN109284363B (en) | 2018-12-03 | 2018-12-03 | Question answering method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811465133.0A CN109284363B (en) | 2018-12-03 | 2018-12-03 | Question answering method and device, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109284363A true CN109284363A (en) | 2019-01-29 |
| CN109284363B CN109284363B (en) | 2023-03-14 |
Family
ID=65174472
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811465133.0A Active CN109284363B (en) | 2018-12-03 | 2018-12-03 | Question answering method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109284363B (en) |
Cited By (55)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109872816A (en) * | 2019-02-13 | 2019-06-11 | 广州天鹏计算机科技有限公司 | The chronic obstructive disease of lung consultation method of knowledge based map, device and system |
| CN109902161A (en) * | 2019-02-01 | 2019-06-18 | 出门问问信息科技有限公司 | Answer processing method, device, equipment and the storage medium of question answering system |
| CN109933660A (en) * | 2019-03-25 | 2019-06-25 | 广东石油化工学院 | The API information search method based on handout and Stack Overflow towards natural language form |
| CN110008326A (en) * | 2019-04-01 | 2019-07-12 | 苏州思必驰信息科技有限公司 | Knowledge abstraction generating method and system in conversational system |
| CN110008413A (en) * | 2019-03-14 | 2019-07-12 | 海信集团有限公司 | A kind of traffic trip problem querying method and device |
| CN110059231A (en) * | 2019-04-19 | 2019-07-26 | 话媒(广州)科技有限公司 | A kind of generation method and device of reply content |
| CN110097970A (en) * | 2019-06-26 | 2019-08-06 | 北京康健数字化健康管理研究院 | A kind of facial paralysis diagnostic system and its system method for building up based on deep learning |
| CN110321437A (en) * | 2019-05-27 | 2019-10-11 | 腾讯科技（深圳）有限公司 | A kind of corpus data processing method, device, electronic equipment and medium |
| CN110427470A (en) * | 2019-07-25 | 2019-11-08 | 腾讯科技（深圳）有限公司 | Question and answer processing method, device and electronic equipment |
| CN110472034A (en) * | 2019-08-21 | 2019-11-19 | 北京百度网讯科技有限公司 | Detection method, device, equipment and the computer readable storage medium of question answering system |
| CN110750633A (en) * | 2019-10-22 | 2020-02-04 | 苏州思必驰信息科技有限公司 | Method and device for determining answer of question |
| CN110866096A (en) * | 2019-10-15 | 2020-03-06 | 平安科技（深圳）有限公司 | Intelligent answer control method and device, computer equipment and storage medium |
| CN110909159A (en) * | 2019-09-29 | 2020-03-24 | 珠海格力电器股份有限公司 | User intention identification method, device, terminal and storage medium |
| CN111125379A (en) * | 2019-12-26 | 2020-05-08 | 科大讯飞股份有限公司 | Knowledge base expansion method and device, electronic equipment and storage medium |
| CN111125150A (en) * | 2019-12-26 | 2020-05-08 | 成都航天科工大数据研究院有限公司 | Industrial field question-answering system retrieval method |
| CN111125361A (en) * | 2019-12-20 | 2020-05-08 | 天津幸福生命科技有限公司 | Data enhancement method and device, storage medium and electronic equipment |
| CN111159381A (en) * | 2019-12-31 | 2020-05-15 | 中国银行股份有限公司 | Data searching method and device |
| CN111198941A (en) * | 2020-01-03 | 2020-05-26 | 联想(北京)有限公司 | Problem discovery method and device, electronic equipment and storage medium |
| CN111291158A (en) * | 2020-01-22 | 2020-06-16 | 北京猎户星空科技有限公司 | Information query method and device, electronic equipment and storage medium |
| CN111368093A (en) * | 2020-02-26 | 2020-07-03 | 京东方科技集团股份有限公司 | Information acquisition method and device, electronic equipment and computer readable storage medium |
| CN111368048A (en) * | 2020-02-26 | 2020-07-03 | 京东方科技集团股份有限公司 | Information acquisition method and device, electronic equipment and computer readable storage medium |
| CN111597162A (en) * | 2020-05-19 | 2020-08-28 | 北京明略软件系统有限公司 | Information determination method and apparatus, storage medium, and electronic apparatus |
| CN111611419A (en) * | 2019-02-26 | 2020-09-01 | 阿里巴巴集团控股有限公司 | Sub-graph identification method and device |
| CN111625640A (en) * | 2020-06-11 | 2020-09-04 | 腾讯科技（深圳）有限公司 | Question and answer processing method, device and storage medium |
| CN111666389A (en) * | 2020-04-21 | 2020-09-15 | 文思海辉智科科技有限公司 | Dialogue data processing method, device, computer equipment and storage medium |
| CN111666388A (en) * | 2020-04-21 | 2020-09-15 | 文思海辉智科科技有限公司 | Dialogue data processing method, device, computer equipment and storage medium |
| CN111739541A (en) * | 2019-03-19 | 2020-10-02 | 上海云思智慧信息技术有限公司 | Conference assistance method and system based on voice, storage medium and terminal |
| CN111737432A (en) * | 2020-06-22 | 2020-10-02 | 四川长虹电器股份有限公司 | Automatic dialogue method and system based on joint training model |
| CN111858877A (en) * | 2020-06-17 | 2020-10-30 | 平安科技（深圳）有限公司 | Multi-type question intelligent question answering method, system, equipment and readable storage medium |
| CN111915339A (en) * | 2019-05-09 | 2020-11-10 | 阿里巴巴集团控股有限公司 | Data processing method, device and equipment |
| CN112035636A (en) * | 2020-08-28 | 2020-12-04 | 康键信息技术（深圳）有限公司 | Question-answer management method, device, equipment and storage medium of medical inquiry system |
| CN112363923A (en) * | 2020-11-09 | 2021-02-12 | 中国平安人寿保险股份有限公司 | Test method, device, computer equipment and medium based on questionnaire system |
| CN112395885A (en) * | 2020-11-27 | 2021-02-23 | 安徽迪科数金科技有限公司 | Short text semantic understanding template generation method, semantic understanding processing method and device |
| CN112685542A (en) * | 2019-10-17 | 2021-04-20 | 阿里巴巴集团控股有限公司 | Method, device, computer system and readable storage medium for question answering system |
| CN112749259A (en) * | 2019-10-29 | 2021-05-04 | 阿里巴巴集团控股有限公司 | Commodity question and answer generation method and device and computer storage medium |
| CN112765194A (en) * | 2020-12-31 | 2021-05-07 | 科大讯飞股份有限公司 | Data retrieval method and electronic equipment |
| CN112800188A (en) * | 2019-11-13 | 2021-05-14 | 阿里巴巴集团控股有限公司 | Conversation processing method and device |
| CN112836061A (en) * | 2021-01-12 | 2021-05-25 | 平安科技（深圳）有限公司 | Intelligent recommendation method and device and computer equipment |
| CN112860865A (en) * | 2021-02-10 | 2021-05-28 | 达而观信息科技（上海）有限公司 | Method, device, equipment and storage medium for realizing intelligent question answering |
| CN112966119A (en) * | 2021-02-25 | 2021-06-15 | 青岛海信网络科技股份有限公司 | Information acquisition method, equipment and medium |
| CN112988784A (en) * | 2021-04-26 | 2021-06-18 | 广州思迈特软件有限公司 | Data query method, query statement generation method and device |
| CN112989005A (en) * | 2021-04-16 | 2021-06-18 | 重庆中国三峡博物馆 | Knowledge graph common sense question-answering method and system based on staged query |
| CN113157863A (en) * | 2021-04-25 | 2021-07-23 | 平安科技（深圳）有限公司 | Question and answer data processing method and device, computer equipment and storage medium |
| CN113449117A (en) * | 2021-06-24 | 2021-09-28 | 武汉工程大学 | Bi-LSTM and Chinese knowledge graph-based composite question-answering method |
| CN113468894A (en) * | 2021-07-27 | 2021-10-01 | 泰康保险集团股份有限公司 | Dialogue interaction method and device, electronic equipment and computer-readable storage medium |
| CN113641814A (en) * | 2021-10-18 | 2021-11-12 | 广州讯鸿网络技术有限公司 | Intelligent robot communication system, method and device based on 5G message |
| CN113987145A (en) * | 2021-10-22 | 2022-01-28 | 智联(无锡)信息技术有限公司 | Method, system, equipment and storage medium for accurately reasoning user attribute entity |
| CN114328959A (en) * | 2021-12-27 | 2022-04-12 | 北京百度网讯科技有限公司 | Knowledge graph construction and use method, device, equipment and medium |
| CN115576435A (en) * | 2022-12-12 | 2023-01-06 | 深圳市人马互动科技有限公司 | Intention processing method and related device |
| CN116186232A (en) * | 2023-04-26 | 2023-05-30 | 中国电子技术标准化研究院 | Standard knowledge intelligent question-answering implementation method, device, equipment and medium |
| CN116881426A (en) * | 2023-08-30 | 2023-10-13 | 环球数科集团有限公司 | AIGC-based self-explanatory question-answering system |
| CN116955573A (en) * | 2023-09-19 | 2023-10-27 | 北京奇虎科技有限公司 | Question searching method, device, equipment and storage medium |
| CN117077792A (en) * | 2023-10-12 | 2023-11-17 | 支付宝(杭州)信息技术有限公司 | Knowledge graph-based method and device for generating prompt data |
| CN117556920A (en) * | 2023-10-23 | 2024-02-13 | 星环信息科技(上海)股份有限公司 | Large model illusion treatment method, device, equipment and storage medium |
| CN118277588A (en) * | 2024-05-30 | 2024-07-02 | 阿里云飞天(杭州)云计算技术有限公司 | Query request processing method, electronic device and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106095932A (en) * | 2016-06-13 | 2016-11-09 | 竹间智能科技（上海）有限公司 | Encyclopaedic knowledge question sentence recognition methods and device |
| CN106919655A (en) * | 2017-01-24 | 2017-07-04 | 网易（杭州）网络有限公司 | A kind of answer provides method and apparatus |
| US20180285346A1 (en) * | 2017-03-31 | 2018-10-04 | Baidu Online Network Technology (Beijing) Co., Ltd. | Method and apparatus for updating mining model |
-
2018
- 2018-12-03 CN CN201811465133.0A patent/CN109284363B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106095932A (en) * | 2016-06-13 | 2016-11-09 | 竹间智能科技（上海）有限公司 | Encyclopaedic knowledge question sentence recognition methods and device |
| CN106919655A (en) * | 2017-01-24 | 2017-07-04 | 网易（杭州）网络有限公司 | A kind of answer provides method and apparatus |
| US20180285346A1 (en) * | 2017-03-31 | 2018-10-04 | Baidu Online Network Technology (Beijing) Co., Ltd. | Method and apparatus for updating mining model |
Cited By (91)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109902161A (en) * | 2019-02-01 | 2019-06-18 | 出门问问信息科技有限公司 | Answer processing method, device, equipment and the storage medium of question answering system |
| CN109902161B (en) * | 2019-02-01 | 2023-10-20 | 出门问问创新科技有限公司 | Answer processing method, device, equipment and storage medium of question-answering system |
| CN109872816A (en) * | 2019-02-13 | 2019-06-11 | 广州天鹏计算机科技有限公司 | The chronic obstructive disease of lung consultation method of knowledge based map, device and system |
| CN111611419B (en) * | 2019-02-26 | 2023-06-20 | 阿里巴巴集团控股有限公司 | Sub-graph identification method and device |
| CN111611419A (en) * | 2019-02-26 | 2020-09-01 | 阿里巴巴集团控股有限公司 | Sub-graph identification method and device |
| CN110008413A (en) * | 2019-03-14 | 2019-07-12 | 海信集团有限公司 | A kind of traffic trip problem querying method and device |
| CN111739541A (en) * | 2019-03-19 | 2020-10-02 | 上海云思智慧信息技术有限公司 | Conference assistance method and system based on voice, storage medium and terminal |
| CN109933660A (en) * | 2019-03-25 | 2019-06-25 | 广东石油化工学院 | The API information search method based on handout and Stack Overflow towards natural language form |
| CN110008326A (en) * | 2019-04-01 | 2019-07-12 | 苏州思必驰信息科技有限公司 | Knowledge abstraction generating method and system in conversational system |
| CN110008326B (en) * | 2019-04-01 | 2020-11-03 | 苏州思必驰信息科技有限公司 | Knowledge abstract generation method and system in session system |
| CN110059231A (en) * | 2019-04-19 | 2019-07-26 | 话媒(广州)科技有限公司 | A kind of generation method and device of reply content |
| CN111915339A (en) * | 2019-05-09 | 2020-11-10 | 阿里巴巴集团控股有限公司 | Data processing method, device and equipment |
| CN110321437B (en) * | 2019-05-27 | 2024-03-15 | 腾讯科技（深圳）有限公司 | Corpus data processing method and device, electronic equipment and medium |
| CN110321437A (en) * | 2019-05-27 | 2019-10-11 | 腾讯科技（深圳）有限公司 | A kind of corpus data processing method, device, electronic equipment and medium |
| CN110097970A (en) * | 2019-06-26 | 2019-08-06 | 北京康健数字化健康管理研究院 | A kind of facial paralysis diagnostic system and its system method for building up based on deep learning |
| CN110427470B (en) * | 2019-07-25 | 2024-05-28 | 腾讯科技（深圳）有限公司 | Question and answer processing method and device and electronic equipment |
| CN110427470A (en) * | 2019-07-25 | 2019-11-08 | 腾讯科技（深圳）有限公司 | Question and answer processing method, device and electronic equipment |
| CN110472034A (en) * | 2019-08-21 | 2019-11-19 | 北京百度网讯科技有限公司 | Detection method, device, equipment and the computer readable storage medium of question answering system |
| CN110472034B (en) * | 2019-08-21 | 2022-11-15 | 北京百度网讯科技有限公司 | Detection method, device and equipment of question-answering system and computer readable storage medium |
| CN110909159A (en) * | 2019-09-29 | 2020-03-24 | 珠海格力电器股份有限公司 | User intention identification method, device, terminal and storage medium |
| CN110909159B (en) * | 2019-09-29 | 2022-09-16 | 珠海格力电器股份有限公司 | User intention identification method, device, terminal and storage medium |
| CN110866096A (en) * | 2019-10-15 | 2020-03-06 | 平安科技（深圳）有限公司 | Intelligent answer control method and device, computer equipment and storage medium |
| CN112685542B (en) * | 2019-10-17 | 2024-02-20 | 阿里巴巴集团控股有限公司 | Method, apparatus, computer system and readable storage medium for question-answering system |
| CN112685542A (en) * | 2019-10-17 | 2021-04-20 | 阿里巴巴集团控股有限公司 | Method, device, computer system and readable storage medium for question answering system |
| CN110750633A (en) * | 2019-10-22 | 2020-02-04 | 苏州思必驰信息科技有限公司 | Method and device for determining answer of question |
| CN110750633B (en) * | 2019-10-22 | 2022-03-29 | 思必驰科技股份有限公司 | Method and device for determining answer of question |
| CN112749259B (en) * | 2019-10-29 | 2024-01-02 | 阿里巴巴集团控股有限公司 | Commodity question and answer generation method and device and computer storage medium |
| CN112749259A (en) * | 2019-10-29 | 2021-05-04 | 阿里巴巴集团控股有限公司 | Commodity question and answer generation method and device and computer storage medium |
| CN112800188A (en) * | 2019-11-13 | 2021-05-14 | 阿里巴巴集团控股有限公司 | Conversation processing method and device |
| CN112800188B (en) * | 2019-11-13 | 2024-02-27 | 阿里巴巴集团控股有限公司 | Dialogue processing method and device |
| CN111125361A (en) * | 2019-12-20 | 2020-05-08 | 天津幸福生命科技有限公司 | Data enhancement method and device, storage medium and electronic equipment |
| CN111125361B (en) * | 2019-12-20 | 2024-02-02 | 北京懿医云科技有限公司 | Data enhancement method and device, storage medium and electronic equipment |
| CN111125379A (en) * | 2019-12-26 | 2020-05-08 | 科大讯飞股份有限公司 | Knowledge base expansion method and device, electronic equipment and storage medium |
| CN111125150A (en) * | 2019-12-26 | 2020-05-08 | 成都航天科工大数据研究院有限公司 | Industrial field question-answering system retrieval method |
| CN111125379B (en) * | 2019-12-26 | 2022-12-06 | 科大讯飞股份有限公司 | Knowledge base expansion method and device, electronic equipment and storage medium |
| CN111125150B (en) * | 2019-12-26 | 2023-12-26 | 成都航天科工大数据研究院有限公司 | Search method for industrial field question-answering system |
| CN111159381A (en) * | 2019-12-31 | 2020-05-15 | 中国银行股份有限公司 | Data searching method and device |
| CN111159381B (en) * | 2019-12-31 | 2023-05-26 | 中国银行股份有限公司 | Data searching method and device |
| CN111198941A (en) * | 2020-01-03 | 2020-05-26 | 联想(北京)有限公司 | Problem discovery method and device, electronic equipment and storage medium |
| CN111291158B (en) * | 2020-01-22 | 2023-06-09 | 北京猎户星空科技有限公司 | Information query method and device, electronic equipment and storage medium |
| CN111291158A (en) * | 2020-01-22 | 2020-06-16 | 北京猎户星空科技有限公司 | Information query method and device, electronic equipment and storage medium |
| CN111368093B (en) * | 2020-02-26 | 2024-04-30 | 京东方科技集团股份有限公司 | Information acquisition method, information acquisition device, electronic equipment and computer readable storage medium |
| CN111368048A (en) * | 2020-02-26 | 2020-07-03 | 京东方科技集团股份有限公司 | Information acquisition method and device, electronic equipment and computer readable storage medium |
| CN111368093A (en) * | 2020-02-26 | 2020-07-03 | 京东方科技集团股份有限公司 | Information acquisition method and device, electronic equipment and computer readable storage medium |
| CN111666389A (en) * | 2020-04-21 | 2020-09-15 | 文思海辉智科科技有限公司 | Dialogue data processing method, device, computer equipment and storage medium |
| CN111666388B (en) * | 2020-04-21 | 2023-11-10 | 文思海辉智科科技有限公司 | Dialogue data processing method, device, computer equipment and storage medium |
| CN111666388A (en) * | 2020-04-21 | 2020-09-15 | 文思海辉智科科技有限公司 | Dialogue data processing method, device, computer equipment and storage medium |
| CN111597162A (en) * | 2020-05-19 | 2020-08-28 | 北京明略软件系统有限公司 | Information determination method and apparatus, storage medium, and electronic apparatus |
| CN111625640B (en) * | 2020-06-11 | 2023-11-14 | 腾讯科技（深圳）有限公司 | Question and answer processing method, device and storage medium |
| CN111625640A (en) * | 2020-06-11 | 2020-09-04 | 腾讯科技（深圳）有限公司 | Question and answer processing method, device and storage medium |
| WO2021109690A1 (en) * | 2020-06-17 | 2021-06-10 | 平安科技（深圳）有限公司 | Multi-type question smart answering method, system and device, and readable storage medium |
| CN111858877A (en) * | 2020-06-17 | 2020-10-30 | 平安科技（深圳）有限公司 | Multi-type question intelligent question answering method, system, equipment and readable storage medium |
| CN111737432A (en) * | 2020-06-22 | 2020-10-02 | 四川长虹电器股份有限公司 | Automatic dialogue method and system based on joint training model |
| CN112035636A (en) * | 2020-08-28 | 2020-12-04 | 康键信息技术（深圳）有限公司 | Question-answer management method, device, equipment and storage medium of medical inquiry system |
| CN112035636B (en) * | 2020-08-28 | 2023-06-02 | 康键信息技术（深圳）有限公司 | Question and answer management method, device and equipment of medical question and answer system and storage medium |
| CN112363923B (en) * | 2020-11-09 | 2024-05-14 | 中国平安人寿保险股份有限公司 | Testing method and device based on questionnaire system, computer equipment and medium |
| CN112363923A (en) * | 2020-11-09 | 2021-02-12 | 中国平安人寿保险股份有限公司 | Test method, device, computer equipment and medium based on questionnaire system |
| CN112395885B (en) * | 2020-11-27 | 2024-01-26 | 安徽迪科数金科技有限公司 | Short text semantic understanding template generation method, semantic understanding processing method and device |
| CN112395885A (en) * | 2020-11-27 | 2021-02-23 | 安徽迪科数金科技有限公司 | Short text semantic understanding template generation method, semantic understanding processing method and device |
| CN112765194A (en) * | 2020-12-31 | 2021-05-07 | 科大讯飞股份有限公司 | Data retrieval method and electronic equipment |
| CN112765194B (en) * | 2020-12-31 | 2024-04-30 | 科大讯飞股份有限公司 | Data retrieval method and electronic equipment |
| CN112836061A (en) * | 2021-01-12 | 2021-05-25 | 平安科技（深圳）有限公司 | Intelligent recommendation method and device and computer equipment |
| CN112860865A (en) * | 2021-02-10 | 2021-05-28 | 达而观信息科技（上海）有限公司 | Method, device, equipment and storage medium for realizing intelligent question answering |
| CN112966119B (en) * | 2021-02-25 | 2022-11-25 | 青岛海信网络科技股份有限公司 | Information acquisition method, equipment and medium |
| CN112966119A (en) * | 2021-02-25 | 2021-06-15 | 青岛海信网络科技股份有限公司 | Information acquisition method, equipment and medium |
| CN112989005B (en) * | 2021-04-16 | 2022-07-12 | 重庆中国三峡博物馆 | Knowledge graph common sense question-answering method and system based on staged query |
| CN112989005A (en) * | 2021-04-16 | 2021-06-18 | 重庆中国三峡博物馆 | Knowledge graph common sense question-answering method and system based on staged query |
| CN113157863A (en) * | 2021-04-25 | 2021-07-23 | 平安科技（深圳）有限公司 | Question and answer data processing method and device, computer equipment and storage medium |
| CN113157863B (en) * | 2021-04-25 | 2024-05-28 | 平安科技（深圳）有限公司 | Question-answer data processing method, device, computer equipment and storage medium |
| CN112988784A (en) * | 2021-04-26 | 2021-06-18 | 广州思迈特软件有限公司 | Data query method, query statement generation method and device |
| CN113449117B (en) * | 2021-06-24 | 2023-09-26 | 武汉工程大学 | Bi-LSTM and Chinese knowledge graph based compound question-answering method |
| CN113449117A (en) * | 2021-06-24 | 2021-09-28 | 武汉工程大学 | Bi-LSTM and Chinese knowledge graph-based composite question-answering method |
| CN113468894B (en) * | 2021-07-27 | 2024-03-22 | 泰康保险集团股份有限公司 | Dialogue interaction method and device, electronic equipment and computer readable storage medium |
| CN113468894A (en) * | 2021-07-27 | 2021-10-01 | 泰康保险集团股份有限公司 | Dialogue interaction method and device, electronic equipment and computer-readable storage medium |
| CN113641814A (en) * | 2021-10-18 | 2021-11-12 | 广州讯鸿网络技术有限公司 | Intelligent robot communication system, method and device based on 5G message |
| CN113987145B (en) * | 2021-10-22 | 2024-02-02 | 智联网聘信息技术有限公司 | Method, system, equipment and storage medium for accurately reasoning user attribute entity |
| CN113987145A (en) * | 2021-10-22 | 2022-01-28 | 智联(无锡)信息技术有限公司 | Method, system, equipment and storage medium for accurately reasoning user attribute entity |
| CN114328959B (en) * | 2021-12-27 | 2023-03-10 | 北京百度网讯科技有限公司 | Knowledge graph construction and use method, device, equipment and medium |
| CN114328959A (en) * | 2021-12-27 | 2022-04-12 | 北京百度网讯科技有限公司 | Knowledge graph construction and use method, device, equipment and medium |
| CN115576435B (en) * | 2022-12-12 | 2023-04-04 | 深圳市人马互动科技有限公司 | Intention processing method and related device |
| CN115576435A (en) * | 2022-12-12 | 2023-01-06 | 深圳市人马互动科技有限公司 | Intention processing method and related device |
| CN116186232A (en) * | 2023-04-26 | 2023-05-30 | 中国电子技术标准化研究院 | Standard knowledge intelligent question-answering implementation method, device, equipment and medium |
| CN116881426A (en) * | 2023-08-30 | 2023-10-13 | 环球数科集团有限公司 | AIGC-based self-explanatory question-answering system |
| CN116881426B (en) * | 2023-08-30 | 2023-11-10 | 环球数科集团有限公司 | AIGC-based self-explanatory question-answering system |
| CN116955573B (en) * | 2023-09-19 | 2024-03-29 | 北京奇虎科技有限公司 | Question searching method, device, equipment and storage medium |
| CN116955573A (en) * | 2023-09-19 | 2023-10-27 | 北京奇虎科技有限公司 | Question searching method, device, equipment and storage medium |
| CN117077792B (en) * | 2023-10-12 | 2024-01-09 | 支付宝(杭州)信息技术有限公司 | Knowledge graph-based method and device for generating prompt data |
| CN117077792A (en) * | 2023-10-12 | 2023-11-17 | 支付宝(杭州)信息技术有限公司 | Knowledge graph-based method and device for generating prompt data |
| CN117556920A (en) * | 2023-10-23 | 2024-02-13 | 星环信息科技(上海)股份有限公司 | Large model illusion treatment method, device, equipment and storage medium |
| CN117556920B (en) * | 2023-10-23 | 2024-05-31 | 星环信息科技(上海)股份有限公司 | Large model illusion treatment method, device, equipment and storage medium |
| CN118277588A (en) * | 2024-05-30 | 2024-07-02 | 阿里云飞天(杭州)云计算技术有限公司 | Query request processing method, electronic device and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109284363B (en) | 2023-03-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109284363B (en) | Question answering method and device, electronic equipment and storage medium | |
| CN108345690B (en) | Intelligent question and answer method and system | |
| CN107391677B (en) | Method and device for generating Chinese general knowledge graph with entity relation attributes | |
| CN111353310A (en) | Named entity identification method and device based on artificial intelligence and electronic equipment | |
| CN108664599B (en) | Intelligent question-answering method and device, intelligent question-answering server and storage medium | |
| CN109947952B (en) | Retrieval method, device, equipment and storage medium based on English knowledge graph | |
| CN112883165B (en) | Intelligent full-text retrieval method and system based on semantic understanding | |
| CN112328800A (en) | System and method for automatically generating programming specification question answers | |
| CN114153994A (en) | Medical insurance information question-answering method and device | |
| CN115563313A (en) | Knowledge graph-based document book semantic retrieval system | |
| CN116244412A (en) | Multi-intention recognition method and device | |
| CN112507089A (en) | Intelligent question-answering engine based on knowledge graph and implementation method thereof | |
| CN114239828A (en) | Supply chain affair map construction method based on causal relationship | |
| CN114840685A (en) | Emergency plan knowledge graph construction method | |
| Arbaaeen et al. | Natural language processing based question answering techniques: A survey | |
| CN117874204A (en) | Knowledge question-answering method, system, storage medium and computer equipment | |
| CN116049376B (en) | Method, device and system for retrieving and replying information and creating knowledge | |
| CN110750632B (en) | Improved Chinese ALICE intelligent question-answering method and system | |
| CN116450855A (en) | Knowledge graph-based reply generation strategy method and system for question-answering robot | |
| Xie et al. | Research and implementation of automatic question answering system based on ontology | |
| CN115757585A (en) | Enterprise knowledge base data storage method and device and storage medium | |
| CN115982322A (en) | Water conservancy industry design field knowledge graph retrieval method and retrieval system | |
| CN113779981A (en) | Recommendation method and device based on pointer network and knowledge graph | |
| CN111897932A (en) | Query processing method and system for text big data | |
| Arbizu | Extracting knowledge from documents to construct concept maps |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20230612 Address after: 210034 floor 8, building D11, Hongfeng Science Park, Nanjing Economic and Technological Development Zone, Jiangsu Province Patentee after: New Technology Co.,Ltd. Patentee after: VOLKSWAGEN (CHINA) INVESTMENT Co.,Ltd. Address before: 100080 Room 501, 5th floor, NO.67, North Fourth Ring Road West, Haidian District, Beijing Patentee before: Beijing Yufanzhi Information Technology Co.,Ltd. |